Energy from steam heat? What I'm working on...

Sailah

Yeah it was a misleading thread title

Just thought some would be interested in my latest side project of energy harvesting. Our steam trap monitor is predicted to last 5-7 years on a single 3V lithium cell and we feel that's pretty good. But with an abundance of sensors being deployed in industry at some point you are trading 1 maintenance task for the other. So the concept of energy harvesting is interesting to me and I set out to see what's what.

First made a solar powered monitor that recharges a 3.6V Lithium Ion cell. And that works fine, but solar isn't a good option in my opinion due to ambient light available in mechanical spaces. Outdoors yeah but indoors it's tough to get much.


Then I started getting interested in thermal harvesting due to endless supply of heat. The concept that makes this work is a Peltier module. These are small ceramic devices that, when a voltage is applied, transfer heat from one side to the other. Often used to cool computers etc. But in reverse, the Seebeck Effect, creates a voltage and current. In order to do this you need to set up a temperature gradient. Our housing is aluminum so I thermally bonded the Peltier device to the aluminum and put a heat sink on the "hot" side. The "cold" side would be the outer cover of my monitor facing away from the steam trap.



So then I needed to insulate the rest of my housing to prevent it from also absorbing the heat. Easiest solution was some 3mm foam I had and then covering with foil tape. I then placed monitor about same distance away from heat source as it would be in field by hovering over a heat plate.

The wires coming out of my monitor housing are from the Peltier device. They feed into a boost converter DC-DC circuit that is designed to maximize a primary (non rechargeable) battery. So all energy harvested first goes to charging up various capacitors (energy storage).

My left Fluke reads 3.61V which is the reference voltage of the output. So that capacitor is fully charged and ready to power my device when needed. I have another capacitor that essentially stores excess energy in reserve, (Right Fluke @ .72V and hoovering up power from the harvesting process until it also reaches 3.6V).

So when my monitor requires power it will first pull from the output capacitor, and once that drops ~220mV, the storage capacitor kicks in. If that is depleted, it will seamlessly switch to the battery. So in theory, if you are harvesting enough energy, you would never deplete battery and it would run forever. There are 40 year shelf life batteries out there.



If you've stuck with me so far, you're probably asking, yeah OK so how much power can you actually get?

Right now with a 5C temperature delta, I'm getting ~40mV, 12mA and 0.5mW of power. Not sure how to give you context but in 24 hours my monitor uses ~6 Joules of energy and I'm producing ~50 Joules. So way in excess of what I require to run a full spectral analysis of ultrasonic steam trap data and have the ability to send that data via an encrypted packet up to 100 miles line of sight.




Pretty neat what you can do if you have a heat source available. For lower power devices like Bluetooth in the home, you could easily run this from a tiny peltier device.

So if you want to power a sensor from a radiator you could set up a Peltier device on radiator and figure out how to heat sink it properly and send data as long as you had heat.

Jamie Hall

You may be on to something, there, @Sailah !

Sailah

Jamie Hall said:

You may be on to something, there, @Sailah !

Thanks Jamie but the concept of thermal energy harvesting is as old as the hills. I'm using an evaluation board, which is what I have wires all over in pic above, from Analog Devices. It's $175 but is nicely breadboarded out for testing purposes. Integrating the actual circuitry into my design would be much less. Harder to make an insulated blankie for the housing to create temp gradient.

But for the right customer, with monitors 50' off the deck in a piping rack that requires a skyjack to reach...might make sense.

The ROI though is that our monitors last 5-7 years on a $1.50 battery. Yeah it might take someone 5 min to replace battery and that would be hard to offset the cost of harvesting.

There are other applications I'm exploring (not steam) where I'd need faster data and more of it. I can see something like this working well when I'm needing power and can only get 6 months of battery life. And in that application my monitor would be physically touching the component and I'd likely be able to get a much better heat transfer.

Steamhead

Yeah, but how many people actually replace batteries?

@sailah, to quote Andrew Paul- get this patented at once!

Sailah

Steamhead said:

Yeah, but how many people actually replace batteries?

@sailah, to quote Andrew Paul- get this patented at once!

Ha. I have (pending) patented my monitor and a few other things regarding how we do it, specific to finding failed traps.

But energy harvesting, using off the shelf ICs from Analog Devices is likely not going to get a patent. This circuit is probably 8-10 year old technology.

Been doing some more testing this morning and removed battery completely from circuit. So now I am running on thermal energy and supercaps alone. There is a very tiny draw on battery to maintain reference voltage.

Total parasitic battery current draw is 82 nA. So over 1 day I'm using 0.02 joules. My battery has 31,104 joules. Which is 4,229 years. Hopefully my ancestors will be around to change the battery then.

Will we still be using steam in 4000 years? I would think yes. Maybe not for heat but how else are you supposed to eat broccoli?

RPK

Have always wanted to try out these energy harvesting zone valves. Enocean wireless technology is pretty interesting, but I haven’t seen it used with any of the energy harvesting options yet.

http://www.automatedbuildings.com/news/mar12/articles/spartan/120220031202spartan.html